In any of audio playback, you have two major industries involved: the recording industry that record/makes the source and the hardware manufacturers producing the hardware. Then there is psychoacoustics trying to explain what we can hear and how we can hear. In the early days of recording industry, the purpose is tried to record the sound of great performance. I have a friend who was seriously into great classic performers. He gave me this so-called "great performance" filled with hiss and pops and worse of all, the sound was warbling. I never was able to listen to them for more than 10 seconds. Then in the 70's (I guess), the recording industry takes a turn and begin to make recording with good acoustic effects. So they pick church and other similar places with huge sound stage. The recording made in studio is actually too dry in comparison. Later on, DSP catch on and try to emulate that. I have a Denon 4800 and they have simulation mode. Mostly what it does is to add reverberation. To demonstrate this, I create a impulse with a gradual roll-off at 200hz (One of the problem with impulse is the energy is primary focus on high frequency band and it does not have enough resolution for low frequency bands). This 200hz is try to correct that. So here is the impulse input signal to Denon receiver:
So here is the output of Jazz simulation mode. It just adds a bunch of tiny replicates of original signal in order to emulate the real world reverberation scenario.
This is really the "MIDI" version of what happen in real world. In the real world, that impulse will be replaced with far more complicated music signal and still we want our brain to be able to tell all the subsequent tiny signals are the reverberation of first signals, instead of another fresh new signals. That basically explains the difference between good speakers and less speakers. Less speakers prevent us from connecting the dots and we just hear a bunch of noise in the background. With good speakers, the noise can be connected as correlated to each other and that helps us to build a good perceived sound stage. That is why I have emphasized the importance of memory effect in speakers, make sure speakers are not "timing variant systems".
One thing worth noting is there is extra energy associated with the reverberations. So if we do a FFT frequency plot, there should be frequency band got boosted. So we next look at the FFT frequency plot. First, the source impulse:
Each vertical division is 5db. Then frequency plot of the Jazz DSP simulation output:
It is pretty noisy. So we do a smoothing on the plot and overlay the before and after.
Smoothing is a powerful tool. A lot of time, it also wipes out detail that we should pay attention to. But in our case, we just want to find out where the extra energy go. We need to overlay the source FFT plot and this new smoothed FFT plot.
The plot may look a bit different because I increase each vertical to 10db. Note, the low end roll-off is due to AC coupling in Denon receiver. The boost is all at the bass band. In short, if we want to capture the impact of reverberation, we need to have good bass extension. What if we have limited bass extension, say 100hz, Q=1.1 2nd HPF system? Here is the time domain response:
What if we have a 15hz, Q=0.5 bass extension? Here is the time domain response.
If we increase the Q value to say Bessel, here is the response.
Now with Q value even higher to 1.1.
The above also has an implication on bass EQ. We should note our in-room bass response is naturally higher. So if we EQ it down to match the midrange, we essentially bring down lower bass component (as compared with the upper bass) in the first arrival wave. It destroys the spectral balance in the first arrival wave. That may not be a wise thing to do. I normally run my sub 3-6db hotter.
One thing I need to emphasize I called it "MIDI" grade because we have to simply this signal so such so we can visually understand it. But what is lost is I cannot really tell no difference of playing back on one subwoofer vs another with these MIDI grade signals. So speaking of MIDI signals, I receive an email from another AVS member which I think timing cannot be better.
"Before I ask my questions, I need to mention that I have owned my F15 subwoofer for just under one year now (order #3725597) and it has changed my appreciation for music in two ways: first, it has seriously affected my taste in music. I used to spend most of my time listening to modern electronic music, but since adding the F15 to my system, I now enjoy a much wider spectrum of music. Two of my favorites are Donald Fagen and Dire Straits. Second, I was originally under the belief that I would probably need two subwoofers to keep up with my front monitors at a reasonable level of accuracy. I quickly realized that my thinking was backwards – I actually needed to upgrade my front monitors to keep up with the Rythmik."Edited by Rythmik - 8/11/13 at 8:59pm